Multicast overview. Introduction to multicast. Information transmission techniques. Unicast

Transcription

1 Contents Multicast overview 1 Introduction to multicast 1 Information transmission techniques 1 Multicast features 3 Common notations in multicast 4 Multicast advantages and applications 4 Multicast models 5 Multicast architecture 5 Multicast addresses 6 Multicast protocols 9 Multicast packet forwarding mechanism 11 Multicast support for VPNs 11 Introduction to VPN instances 12 Multicast application in VPNs 12 i

2 Multicast overview Introduction to multicast As a technique that coexists with unicast and broadcast, the multicast technique effectively addresses the issue of point-to-multipoint data transmission. By enabling high-efficiency point-to-multipoint data transmission over a network, multicast greatly saves network bandwidth and reduces network load. By using multicast technology, a network operator can easily provide new value-added services, such as live webcasting, web TV, distance learning, telemedicine, web radio, real-time video conferencing, and other bandwidth-critical and time-critical information services. Unless otherwise stated, the term "multicast" in this document refers to IP multicast. The term "router" in this document refers to both routers and Layer 3 switches. Information transmission techniques Unicast The information transmission techniques include unicast, broadcast, and multicast. In unicast transmission, the information source must send a separate copy of information to each host that needs the information. Figure 1 Unicast transmission Host A Receiver Host B Source Host C Receiver Host D IP network Packets for Host B Packets for Host D Packets for Host E Receiver Host E In Figure 1, assume that Host B, Host D, and Host E need the information. A separate transmission channel must be established from the information source to each of these hosts. 1

3 Broadcast In unicast transmission, the traffic transmitted over the network is proportional to the number of hosts that need the information. If a large number of hosts need the information, the information source must send a separate copy of the same information to each of these hosts. Sending many copies can place a tremendous pressure on the information source and the network bandwidth. Unicast is not suitable for batch transmission of information. In broadcast transmission, the information source sends information to all hosts on the subnet, even if some hosts do not need the information. Figure 2 Broadcast transmission Multicast In Figure 2, assume that only Host B, Host D, and Host E need the information. If the information is broadcast to the subnet, Host A and Host C also receive it. In addition to information security issues, broadcasting to hosts that do not need the information also causes traffic flooding on the same subnet. Broadcast is disadvantageous in transmitting data to specific hosts. Moreover, broadcast transmission is a significant waste of network resources. Unicast and broadcast techniques cannot provide point-to-multipoint data transmissions with the minimum network consumption. Multicast transmission can solve this problem. When some hosts on the network need multicast information, the information sender, or multicast source, sends only one copy of the information. Multicast distribution trees are built through multicast routing protocols, and the packets are replicated only on nodes where the trees branch. 2

4 Figure 3 Multicast transmission The multicast source sends only one copy of the information to a multicast group. Host B, Host D and Host E, which are receivers of the information, must join the multicast group. The routers on the network duplicate and forward the information based on the distribution of the group members. Finally, the information is correctly delivered to Host B, Host D, and Host E. To summarize, multicast has the following advantages: Advantages over unicast Because multicast traffic flows to the farthest-possible node from the source before it is replicated and distributed, an increase in the number of hosts will not increase the load of the source and will not remarkably add to the usage of network resources. Advantages over broadcast Because multicast data is sent only to the receivers that need it, multicast uses network bandwidth reasonably and enhances network security. In addition, data broadcast is confined to the same subnet, but multicast is not. Multicast features Multicast transmission has the following features: A multicast group is a multicast receiver set identified by an IP multicast address. Hosts join a multicast group to become members of the multicast group, before they can receive the multicast data addressed to that multicast group. Typically, a multicast source does not need to join a multicast group. An information sender is called a "multicast source." A multicast source can send data to multiple multicast groups at the same time, and multiple multicast sources can send data to the same multicast group at the same time. All hosts that have joined a multicast group become members of the multicast group. The group memberships are dynamic. Hosts can join or leave multicast groups at any time. Multicast groups are not subject to geographic restrictions. Routers or Layer 3 switches that support Layer 3 multicast are called "multicast routers" or "Layer 3 multicast devices." In addition to providing the multicast routing function, a multicast router can also 3

5 manage multicast group memberships on stub subnets with attached group members. A multicast router itself can be a multicast group member. For a better understanding of the multicast concept, you can compare multicast transmission to the transmission of TV programs. Table 1 Comparing TV program transmission and multicast transmission TV transmission A TV station transmits a TV program through a channel. A user tunes the TV set to the channel. The user starts to watch the TV program transmitted by the TV station through the channel. The user turns off the TV set or tunes to another channel. Multicast transmission A multicast source sends multicast data to a multicast group. A receiver joins the multicast group. The receiver starts to receive the multicast data that the source is sending to the multicast group. The receiver leaves the multicast group or joins another group. Common notations in multicast The following notations are commonly used in multicast transmission: (*, G) Rendezvous point tree (RPT), or a multicast packet that any multicast source sends to multicast group G. Here, the asterisk represents any multicast source, and "G" represents a specific multicast group. (S, G) Shortest path tree (SPT), or a multicast packet that multicast source S sends to multicast group G. Here, "S" represents a specific multicast source, and "G" represents a specific multicast group. For more information about the concepts RPT and SPT, see "Configuring PIM" and "Configuring IPv6 PIM." Multicast advantages and applications Multicast advantages Enhanced efficiency Reduces the CPU load of information source servers and network devices. Optimal performance Reduces redundant traffic. Distributed application Enables point-to-multipoint applications at the cost of minimum network resources. Multicast applications The scenarios in which the multicast technique can be effectively applied are: Multimedia and streaming applications, such as web TV, web radio, and real-time video/audio conferencing Communication for training and cooperative operations, such as distance learning and telemedicine. Data warehouse and financial applications (stock quotes) Any other point-to-multipoint application for data distribution 4

6 Multicast models ASM model SFM model SSM model Based on how the receivers treat the multicast sources, the multicast models include any-source multicast (ASM), source-filtered multicast (SFM), and source-specific multicast (SSM). In the ASM model, any sender can send information to a multicast group as a multicast source, and receivers can join a multicast group identified by a group address, and obtain multicast information addressed to that multicast group. In this model, receivers do not know the positions of the multicast sources in advance. However, they can join or leave the multicast group at any time. The SFM model is derived from the ASM model. To a sender, the two models appear to have the same multicast membership architecture. The SFM model functionally extends the ASM model. The upper-layer software checks the source address of received multicast packets and permits or denies multicast traffic from specific sources. Therefore, receivers can receive the multicast data from only part of the multicast sources. To a receiver, multicast sources are not all valid; they are filtered. Users might be interested in the multicast data from only certain multicast sources. The SSM model provides a transmission service that enables users to specify the multicast sources that they are interested in at the client side. The main difference between the SSM model and the ASM model is that in the SSM model, receivers have already determined the locations of the multicast sources by some other means. In addition, the SSM model uses a multicast address range that is different from that of the ASM/SFM model, and dedicated multicast forwarding paths are established between receivers and the specified multicast sources. Multicast architecture IP multicast addresses the following issues: Where should the multicast source transmit information to? (Multicast addressing.) What receivers exist on the network? (Host registration.) Where is the multicast source that will provide data to the receivers? (Multicast source discovery.) How should information be transmitted to the receivers? (Multicast routing.) IP multicast is an end-to-end service. The multicast architecture involves the following parts: Addressing mechanism A multicast source sends information to a group of receivers through a multicast address. Host registration Receiver hosts can join and leave multicast groups dynamically. This mechanism is the basis for management of group memberships. Multicast routing A multicast distribution tree (a forwarding path tree for multicast data on the network) is constructed for delivering multicast data from a multicast source to receivers. 5

7 Multicast applications A software system that supports multicast applications, such as video conferencing, must be installed on multicast sources and receiver hosts. The TCP/IP stack must support reception and transmission of multicast data. Multicast addresses IP multicast addresses IPv4 multicast addresses: Internet Assigned Numbers Authority (IANA) assigned the Class D address space ( to ) to IPv4 multicast. Table 2 Class D IP address blocks and description Address block to to to Description Reserved permanent group addresses. The IP address is reserved. Other IP addresses can be used by routing protocols and for topology searching, protocol maintenance, and so on. Table 3 lists common permanent group addresses. A packet destined for an address in this block will not be forwarded beyond the local subnet regardless of the TTL value in the IP header. Globally scoped group addresses. This block includes the following types of designated group addresses: /8 SSM group addresses /8 Glop group addresses. Administratively scoped multicast addresses. These addresses are considered locally unique rather than globally unique, and can be reused in domains administered by different organizations without causing conflicts. For more information, see RFC NOTE: "Glop" is a mechanism for assigning multicast addresses between different ASs. By filling an AS number into the middle two bytes of , you get 255 multicast addresses for that AS. For more information, see RFC Table 3 Some reserved multicast addresses Address Description All systems on this subnet, including hosts and routers All multicast routers on this subnet Unassigned DVMR routers OSPF routers OSPF designated routers/backup designated routers ST routers. 6

8 Address Description ST hosts RIPv2 routers Mobile agents DHCP server/relay agent All Protocol Independent Multicast (PIM) routers RSVP encapsulation CBT routers Designated SBM All SBMs VRRP. IPv6 multicast addresses: Figure 4 IPv6 multicast format The meanings of the fields of an IPv6 multicast address are as follows: 0xFF If the most significant eight bits are , this address is an IPv6 multicast address. Flags Contains four bits. Figure 5 Flags field format Table 4 Description on the bits of the Flags field Bit Description 0 Reserved, set to 0. R P When set to 0, this address is an IPv6 multicast address without an embedded RP address. When set to 1, this address is an IPv6 multicast address with an embedded RP address. (The P and T bits must also be set to 1.) When set to 0, this address is an IPv6 multicast address not based on a unicast prefix. When set to 1, this address is an IPv6 multicast address based on a unicast prefix. (The T bit must also be set to 1.) 7

9 Bit T Description When set to 0, this address is an IPv6 multicast address permanently-assigned by IANA. When set to 1, this address is a transient, or dynamically assigned IPv6 multicast address. Scope Contains four bits, which represent the scope of the IPv6 internetwork for which the multicast traffic is intended. Table 5 Scope field values Value Meaning 0, 3, F Reserved 1 Interface-local scope 2 Link-local scope 4 Admin-local scope 5 Site-local scope 6, 7, 9 through D Unassigned 8 Organization-local scope E Global scope Group ID Contains 112 bits. It uniquely identifies an IPv6 multicast group in the scope that the Scope field defines. Ethernet multicast MAC addresses A multicast MAC address identifies a group of receivers at the data link layer. IPv4 multicast MAC addresses: As defined by IANA, the most significant 24 bits of an IPv4 multicast MAC address are 0x01005E. Bit 25 is 0, and the other 23 bits are the least significant 23 bits of a multicast IPv4 address. Figure 6 IPv4-to-MAC address mapping The most significant four bits of a multicast IPv4 address are 1110, so this address is a multicast address. Only 23 bits of the remaining 28 bits are mapped to a MAC address, so 5 bits of the multicast IPv4 address are lost. As a result, 32 multicast IPv4 addresses map to the same IPv4 multicast MAC address. Therefore, in Layer 2 multicast forwarding, a device might receive some multicast data destined for other IPv4 multicast groups. The upper layer must filter such redundant data. 8

10 IPv6 multicast MAC addresses: The most significant 16 bits of an IPv6 multicast MAC address are 0x3333. The least significant 32 bits are the least significant 32 bits of a multicast IPv6 address. Figure 7 An example of IPv6-to-MAC address mapping Multicast protocols Generally, Layer 3 multicast refers to IP multicast working at the network layer. The corresponding multicast protocols are Layer 3 multicast protocols, which include IGMP, MLD, PIM, IPv6 PIM, MSDP, MBGP, IPv6 MBGP. Layer 2 multicast refers to IP multicast working at the data link layer. The corresponding multicast protocols are Layer 2 multicast protocols, which include IGMP snooping, MLD snooping, PIM snooping, IPv6 PIM snooping, multicast VLAN, and IPv6 multicast VLAN. IGMP snooping, PIM snooping, multicast VLAN, IGMP, PIM, MSDP, and MBGP are for IPv4, and MLD snooping, IPv6 PIM snooping, IPv6 multicast VLAN, MLD, IPv6 PIM, and IPv6 MBGP are for IPv6. This section provides only general descriptions about applications and functions of the Layer 2 and Layer 3 multicast protocols in a network. For more information about these protocols, see the related chapters. Layer 3 multicast protocols Layer 3 multicast protocols include multicast group management protocols and multicast routing protocols. Figure 8 Positions of Layer 3 multicast protocols 9

11 Multicast group management protocols: Typically, the internet group management protocol (IGMP) or multicast listener discovery protocol (MLD) is used between hosts and Layer 3 multicast devices that directly connect to the hosts. These protocols define the mechanism of establishing and maintaining group memberships between hosts and Layer 3 multicast devices. Multicast routing protocols: A multicast routing protocol runs on Layer 3 multicast devices to establish and maintain multicast routes and forward multicast packets correctly and efficiently. Multicast routes constitute loop-free data transmission paths (also known as multicast distribution trees) from a data source to multiple receivers. In the ASM model, multicast routes include intra-domain routes and inter-domain routes. An intra-domain multicast routing protocol discovers multicast sources and builds multicast distribution trees within an AS to deliver multicast data to receivers. Among a variety of mature intra-domain multicast routing protocols, protocol independent multicast (PIM) is most widely used. Based on the forwarding mechanism, PIM has dense mode (often referred to as "PIM-DM"), and sparse mode (often referred to as "PIM-SM"). An inter-domain multicast routing protocol is used for delivery of multicast information between two ASs. So far, mature solutions include Multicast Source Discovery Protocol (MSDP) and Multicast Border Gateway Protocol (MBGP). MSDP propagates multicast source information among different ASs. MBGP is an extension of the Multi-protocol Border Gateway Protocol (MP-BGP) for exchanging multicast routing information among different ASs. For the SSM model, multicast routes are not divided into intra-domain routes and inter-domain routes. Because receivers know the position of the multicast source, channels established through PIM-SM are sufficient for the transport of multicast information. Layer 2 multicast protocols Layer 2 multicast protocols include IGMP snooping, MLD snooping, PIM snooping, IPv6 PIM snooping, multicast VLAN, and IPv6 multicast VLAN. Figure 9 Positions of Layer 2 multicast protocols IGMP snooping and MLD snooping: 10

12 IGMP snooping and MLD snooping are multicast constraining mechanisms that run on Layer 2 devices. They manage and control multicast groups by listening to and analyzing IGMP or MLD messages exchanged between the hosts and Layer 3 multicast devices, effectively controlling the flooding of multicast data in a Layer 2 network. PIM snooping and IPv6 PIM snooping: PIM snooping and IPv6 PIM snooping run on Layer 2 devices. They determine which ports are interested in multicast data by analyzing the received IPv6 PIM messages, and add the ports to a multicast forwarding entry to make sure that multicast data can be forwarded to only the ports that are interested in the data. Multicast VLAN and IPv6 multicast VLAN: In the traditional multicast-on-demand mode, when users in different VLANs on a Layer 2 device need multicast information, the upstream Layer 3 device must forward a separate copy of the multicast data to each VLAN of the Layer 2 device. When the multicast VLAN or IPv6 multicast VLAN feature is enabled on the Layer 2 device, the Layer 3 multicast device sends only one copy of multicast to the multicast VLAN or IPv6 multicast VLAN on the Layer 2 device. This method avoids waste of network bandwidth and extra burden on the Layer 3 device. Multicast packet forwarding mechanism In a multicast model, a multicast source sends information to the host group identified by the multicast group address in the destination address field of IP multicast packets. To deliver multicast packets to receivers located at different positions of the network, multicast routers on the forwarding paths usually need to forward multicast packets that an incoming interface receives to multiple outgoing interfaces. Compared with a unicast model, a multicast model is more complex in the following aspects: To ensure multicast packet transmission in the network, unicast routing tables or multicast routing tables (for example, the MBGP routing table) specially provided for multicast must be used as guidance for multicast forwarding. To process the same multicast information from different peers received on different interfaces of the same device, every multicast packet undergoes a reverse path forwarding (RPF) check on the incoming interface. The result of the RPF check determines whether the packet will be forwarded or discarded. The RPF check mechanism is the basis for most multicast routing protocols to implement multicast forwarding. For more information about the RPF mechanism, see "Configuring multicast routing and forwarding" and "Configuring IPv6 multicast routing and forwarding." Multicast support for VPNs NOTE: Multicast support for VPNs is not available in IPv6 networks. Multicast support for VPNs refers to multicast applied in VPNs. 11

13 Introduction to VPN instances VPNs must be isolated from one another and from the public network. As shown in Figure 10, VPN A and VPN B separately access the public network through PE devices. Figure 10 VPN networking diagram VPN A CE a2 CE b2 PE 2 CE b3 VPN B VPN B CE b1 P CE a1 PE 1 Public network PE 3 CE a3 VPN A VPN A The provider (P) device belongs to the public network. The customer edge (CE) devices belong to their respective VPNs. Each CE device serves its own VPN and maintains only one set of forwarding mechanisms. The provider edge (PE) devices connect to the public network and the VPNs. Each PE device must strictly distinguish the information for different networks and maintain a separate forwarding mechanism for each network. On a PE device, a set of software and hardware that serve the same network forms an instance. Multiple instances can exist on the same PE device, and an instance can reside on different PE devices. On a PE device, the instance for the public network is called the public network instance, and those for VPNs are called VPN instances. Multicast application in VPNs A PE device that supports multicast for VPNs does the following operations: Maintains an independent set of multicast forwarding mechanisms for each VPN, including the multicast protocols, PIM neighbor information, and multicast routing table. In a VPN, the device forwards multicast data based on the forwarding table or routing table for that VPN. Implements the isolation between different VPNs. Implements information exchange and data conversion between the public network and VPNs. As shown in Figure 10, when a multicast source in VPN A sends a multicast stream to a multicast group, only the receivers that belong to both the multicast group and VPN A can receive the multicast stream. The multicast data is multicast both in VPN A and in the public network. 12